Classification for Assessing the Quality of Diagnostic Coronary Angiography

Abstract: Coronary angiography remains the gold standard for diagnosing obstructive coronary artery disease. However, no standardized, objective, and quantitative classification to assess the quality of coronary angiography exists. In the present report, we sought to establish a novel standardized quantitative classification for the quality of coronary angiography, taking into consideration two main parameters: coronary contrast filling and coronary sinus contrast reflux. Intraobserver and interobserver reproducibility performed among 70 angiograms and 9 readers demonstrated excellent and substantial reproducibility, respectively. The proposed classification may be useful in daily practice, clinician training, and clinical trials.

J INVASIVE CARDIOL 2017;29(12):417-420. Epub 2017 August 15.

Key words: angiogram, classification, contrast, quality

Coronary angiography is the gold standard for diagnosing obstructive coronary artery disease (CAD).¹ Nonetheless, image quality can vary greatly, potentially affecting the interpretability and diagnostic accuracy of angiography. In general, an optimal coronary injection will ensure complete opacification of the major coronary arteries and side branches, without excessive (wasted) contrast spillover into the coronary sinuses. Despite being performed in more than 1 million patients per year in the United States,² no standardized, objective, and quantitative classification to assess the quality of coronary angiography exists. A classification scheme to evaluate angiographic quality could prove useful to assess quality measures of coronary imaging in daily practice, might provide educational value for training interventional cardiology fellows, and would allow for the systematic and reproducible evaluation of image quality during clinical studies aimed at improving imaging accuracy or reducing contrast exposure. Previous reports suggested criteria that might be considered important for angiographic quality;^3-14 however, these assessments were based on subjective evaluation without objective quantitative criteria. In light of the lack of a definitive classification, we sought to establish a novel standardized quantitative classification for the quality of coronary angiography.

Classification Description

Two main parameters were taken into consideration for assessment of the quality of coronary angiography (Table 1): (1) coronary contrast filling (directly reflecting image quality); and (2) coronary sinus contrast reflux (reflecting the amount of contrast used, with less being desirable). The coronary contrast filling classification is based on the extent and quality of opacification of the main vessel and side branches, assessing the density of vessel filling compared to the density of contrast within the guiding catheter. The contrast filling scale ranges from grade 0 (no or minimal filling of the coronary vessel, ie, a poor or inappropriate injection) to grade III (completely filling the entire vessel including all major side branches with density equal to the guiding catheter, ie, an optimal injection). Subclassification involves the presence of contrast streaming and the number of side branches opacified (Table 1). Figure 1 illustrates each grade. Obviously, factors independent of the quality of the coronary contrast injection may influence vessel filling, including inadequate catheter engagement, the presence of an occlusive thrombus in the setting of ST-segment elevation myocardial infarction, the presence of obstructive devices (atherectomy device or stent delivery system, etc), or the presence of intraprocedural complications (slow-reflow or no-reflow, distal embolization). These factors must be taken into account when assessing the quality of contrast injection. In general, the baseline and final diagnostic angiograms should be assessed to ensure uniformity and minimize the effect of these variables.

The classification of coronary sinus reflux is based on three parameters: (1) the number of sinuses filled (right, left, and non-coronary) by the reflux of contrast during the peak of injection; (2) the completeness of coronary sinus filling; and (3) the contrast density of sinus filling compared to the contrast density of the guiding catheter. Grading ranges from A (no or minimal coronary sinus reflux, ie, an optimal injection, with no or minimal contrast wasted) to D (complete filling of at least 2 coronary sinuses with density equal to the guiding catheter, ie, a poor or inappropriate injection, with excess contrast use). Figure 2 illustrates each grade. Practically speaking, the classification can be summarized as follows: Grade A = no sinus filled; Grade B = 1 sinus incompletely filled; Grade C = at least 1 sinus completely filled; and Grade D = at least 2 sinuses completely filled. Subclassification has been made to better describe each possibility within each grade.

Table 2 presents a proposed scoring system. The total and average scores should be reported for the baseline and final angiograms, also categorized separately for the left and right coronary artery systems. Table 3 presents factors that could potentially influence the quality of the coronary angiogram, either by decreasing coronary filling or increasing coronary sinus reflux. Vessel diameter (ie, large ectatic vessel), the presence of a chronic total occlusion, especially if located in a very proximal segment, and the presence of concomitant valvular disease (ie, aortic stenosis) are well-known factors that could make adequate and complete opacification of the vessel with minimal contrast reflux challenging. Other factors, such as high body mass index, body habitus (eg, large chest), or operator inexperience may negatively impact the quality of the coronary angiogram. Whether access route (radial vs femoral) is associated with decreased coronary angiographic quality, either by poor catheter engagement or the routine use of smaller catheters (eg, 4 Fr or 5 Fr), is a matter of debate and could be assessed in a comparative quality study using this proposed classification. 

Reproducibility of the proposed classification scheme. To assess interobserver reproducibility of the proposed classification scheme, the Fleiss k statistic value for multiple readers was determined among 9 readers (angiographic core laboratory technicians) independently grading 70 angiograms, using the Magree macro (SAS version 9.2; SAS Institute).^15,16 To assess intraobserver reproducibility, the same 70 angiograms were reanalyzed by the same 9 readers 4 weeks after the first analysis, and a simple k statistic value was determined for each readers. Interobserver reproducibility among the 9 readers was substantial (Table 4), while the intraobserver reproducibility was almost perfect (Table 5). 

Utility of classifying the quality of coronary angiography. By using simple and readily assessable criteria (coronary filling and coronary sinus reflux), the proposed classification is intuitive for interventional cardiologists, and provides metrics for training. Importantly, the classification scheme uses concepts familiar in prior classification and grading systems, such as the TIMI flow grade.^17,18 Paired with other variables, such as the total amount of contrast administered, our classification could be used by cath lab directors to assess the adequacy of coronary angiography, be used in randomized trials to provide imaging feedback to enrolling sites, and eventually be correlated with the occurrence of related complications such as contrast-induced nephropathy.¹⁹ The present classification scheme is being used in the ongoing AVERT clinical trial (NCT01976299) to examine whether a device developed to reduce contrast administration can do so without adversely affecting image quality. 

Conclusion

This novel, simple, and intuitive classification, created to better classify and assess the quality of diagnostic coronary angiography, demonstrated excellent interobserver and intraobserver variability. Pending further validation, its implementation in daily practice and clinical research could help to improve patient care and outcomes for those undergoing coronary angiography.

Acknowledgment. The authors would like to thank Jolanta Bak and Juan M. Iturriza for their help in creating the figures.

References

1.    Scanlon PJ, Faxon DP, Audet A-M, et al. ACC/AHA guidelines for coronary angiography. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Coronary Angiography) developed in collaboration with the Society for Cardiac Angiography and Interventions. J Am Coll Cardiol. 1999;33:1756-1824.

2.    Mozaffarian D, Benjamin EJ, Go AS, et al. Heart disease and stroke statistics – 2015 update: a report from the American Heart Association. Circulation. 2015;131:e29-e322.

3.    Bernardi G, Bar O, Jezewski T, et al. Quality criteria for cardiac images: an update. Radiat Prot Dosimetry. 2008;129:87-90.

4.    Bernardi G, Padovani R, Morocutti G, et al. Quality criteria for cardiac images in diagnostic and interventional cardiology. Br J Radiol. 2001;74:852-855.

5.    Chahoud G, Khoukaz S, El-Shafei A, Azrak E, Bitar S, Kern MJ. Randomized comparison of coronary angiography using 4F catheters: 4F manual versus “Acisted” power injection technique. Catheter Cardiovasc Interv. 2001;53:221-224.

6.    Chodor P, Morawski S, Sulik-Gajda S, et al. Evaluation of the usefulness of coronary catheters and 4 Fr insertion sets for transradial access coronarography in comparison with catheters and 5 Fr sets. Postepy Kardiol Interwencyjnej. 2013;9:332-336.

7.    Danzi GB, Capuano C, Sesana M, et al. A randomised comparison of the use of 4 and 6 French diagnostic catheters: the limits of downsizing. Int J Cardiol. 2001;79:113-117; discussion 117-118.

8.    Gonzalez MA, Ben-Dor I, Gaglia MA Jr, et al. Qualitative comparison of coronary angiograms between 4 French catheters with an advanced cardiovascular injection system and 6 French catheters with manual injection. Catheter Cardiovasc Interv. 2012;79:843-848.

9.    Kaye DM, Stub D, Mak V, Doan T, Duffy SJ. Reducing iodinated contrast volume by manipulating injection pressure during coronary angiography. Catheter Cardiovasc Interv. 2014;83:741-745.

10.    Khoukaz S, Kern MJ, Bitar SR, et al. Coronary angiography using 4 Fr catheters with Acisted power injection: a randomized comparison to 6 Fr manual technique and early ambulation. Catheter Cardiovasc Interv. 2001;52:393-398.

11.    Leape LL, Park RE, Bashore TM, Harrison JK, Davidson CJ, Brook RH. Effect of variability in the interpretation of coronary angiograms on the appropriateness of use of coronary revascularization procedures. Am Heart J. 2000;139:106-113.

12.    Mehta H, Chatterjee T, Windecker S, et al. Four French catheters for diagnostic coronary angiography. Catheter Cardiovasc Interv. 2003;58:275-280.

13.    Partridge J, McGahan G, Causton S, et al. Radiation dose reduction without compromise of image quality in cardiac angiography and intervention with the use of a flat panel detector without an antiscatter grid. Heart. 2006;92:507-510.

14.    Todd DM, Hubner PJ, Hudson N, Sarma J, McCance AJ, Caplin J. Multicentre, prospective, randomized trial of 4 vs 6 French catheters in 410 patients undergoing coronary angiography. Catheter Cardiovasc Interv. 2001;54:269-275.

15.    Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33:159-174.

16.    J. F. Statistical Methods for Rates and Proportions. Wiley; New York, NY. 1981.

17.    The TIMI Study Group. The Thrombolysis in Myocardial Infarction (TIMI) trial. N Engl J Med. 1985;312:932-936.

18.    The TIMI Study Group. Comparison of invasive and conservative strategies after treatment with intravenous tissue plasminogen activator in acute myocardial infarction. N Engl J Med. 1989;320:618-627.

19.    Mehran R, Aymong ED, Nikolsky E, et al. A simple risk score for prediction of contrast-induced nephropathy after percutaneous coronary intervention: development and initial validation. J Am Coll Cardiol. 2004;44:1393-1399.

From the ¹Gagnon Cardiovascular Institute, Morristown Medical Center, Morristown, New Jersey; ²Cardiovascular Research Foundation, New York, New York; ³Hôpital du Sacré-Coeur de Montréal, Université de Montréal, Montréal, Quebec, Canada; ⁴Icahn School of Medicine at Mount Sinai, New York, New York; and ⁵Columbia University Medical Center, New York, New York.

Disclosures: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Mehran reports grants: AstraZeneca, Bayer, Beth Israel Deaconess, BMS, CSL Behring, Eli Lilly/DSI, Novartis Pharmaceuticals, OrbusNeich, and Medtronic; consultant fees to institution: Abbott Laboratories; consultant fees to spouse: Abiomed, The Medicines Company, CardioKinetix, and Spectranetics; personal fees: Boston Scientific, CSI, Shanghai BraccoSine Pharmaceutical, Janssen Pharmaceuticals, Medscape, and Osprey Medical; advisory board fees to institution: BMS; <1% equity in Claret Medical and Elixir Medical; DSMB membership (to institution) by Watermark Research Partners. Dr Stone reports personal fees: St. Jude, Toray, Matrizyme, Ablative Solutions, Claret, Sirtex, Medical Development Technologies, Vascular Dynamics, Miracor, Neovasc, V-wave, BackBeat Medical, Valfix, TherOx, Reva, and Qool Therapeutics; equity in Caliber, Aria, Biostar family of funds, MedFocus family of funds, Guided Delivery Systems, Micardia, and Cagent; employer receives royalties from Abbott Vascular for sale of the MitraClip. The remaining authors report no financial relationships or conflicts of interest regarding the content herein.

Manuscript submitted March 16, 2017, provisional acceptance given March 29, 2017, final version accepted April 4, 2017.

Address for correspondence: Philippe Généreux, MD, Cardiovascular Research Foundation, 1700 Broadway, 9th Floor, New York, NY 10019. Email: pgenereux@crf.org